I didn't measure the difference in energy consumption whilst it's waiting between recordings. I'll do that. 

Having it turn off after a few minutes is a good idea.

The electronic design around the MEMS microphone is pretty much optimised. We found that a lot of the noise is mechanical from capacitors on the board, and in the SD card, physically expanding and contracting as they balance out the voltage due to the larger current draw from the SD card. Some SD cards are much quieter than others, and we always use SanDIsk Extreme cards. The SanDisk Extreme Pro cards has much higher peak currents during write operations. The interference gets a bit stronger as the batteries run down as their increasing internal resistance causes more voltage fluctuation. 

Hey @JakeBurton , nice work with The Inventory! 

This is very aligned with some internal efforts going on in our community to map out devices, but also data and models eventually. It would be great if we find a way to make this two interoperable. We are still in the (very) alpha development phases, so it is quite open, but it would be super helpful to have some insight about the under-the-hood data structure that The Inventory is relying on, so we can figure out good ways to reliably map or info onto them. 

Happy to chat further or set up a meeting! 

@qgeissmann👀

Thanks very much @LucaPego! Just to clarify, 'Products' does cover hardware, but it also includes software and databases as well. Basically anything that you can take and use in your work. Sounds like we should also be including models in that as well!

It would be great to see more databases and software listed and reviewed on there. We would also be interested to learn more about your internal efforts in your community and what overlap there is for sure. Could we set up a meeting?

Hey there @JakeBurton , sorry, I did not see your message! 

Why don't you shoot me an email with some tentative availability to luca.pegoraro (at) wsl.ch, and we take it from there? 

Again sorry for the delay! 

Drones have changed how we collect information and do surveys. Whether you’re new to drones or a pro wanting to use aerial data, knowing how to do a drone survey is essential. 

@sebjwallace Glad you gave iNaturalist a go, and happy to hear you enjoyed it. Now tell all your friends!

Hi Alex,

Very cool, that would give us lots of room for collaboration since we are focussing on model development taking hardware restrictions and inference speed into account.

Honestly, no (at least not for wildlife applications).

I suspect for the manufacturing community it's a balancing act, and I am not confident that the manufacturing community has invested a lot of resources or effort to integrate sustainability in their tag designs specifically related to the materials. Reasons that come to mind are material durability and cost (both materials costs and cost to for R&D). 

With that said, I think Stuart (below) makes a good suggestion...

We are ready to make efforts on materials, but today there are few solutions available. Today, our partners' R&D efforts are focused on renewable energies and the energy source's capacity. In the absence others solutions, and to take action, we have chosen to develop solutions to recover these tags, some of which are reusable (thanks to Gonio RXG-234 @ThomasGray). To date, thousands of tags have been recovered and some reused. This is the "upcycling" of science! It's a small step, but a great leap forward!

There's elk in Santa Clara county?  I thought the nearest herds in the bay area were at the preserve in Point Reyes.

Thanks for sharing - hadn't realized that. 

Yes, this system is designed to be installed near farms. We also have the repeller system with audio & light, that is battery & solar powered. This system is a "last line of defence". The blowers alone requires 1000 watts :)

Oh yeah that would be awesome! Let me email you to follow up.  I assume you're working with Alec Christie then? He was sharing your team's work in our chatgpt discussion: 

Yes,  if the canopy is sparse enough, you can see through the canopy with TIR what you cannot see in the RGB. We had tested with large mammals like rhinos and elephants that we could not see at all with the RGB under a semi-sparse canopy but were very clearly visible in TIR. It was actually quite surprising how easily we could detect the mammals under the canopy. It's likely similar for mid-sized mammals that live in the canopy that those drier seasons will be much easier to detect, although we did not test small mammals for visibility through the seasons. Other research has and there are a number of studies on primates now. 

I did quite a bit of flying above the canopy, and did not have many problems. It's just a matter of always flying bit higher than the canopy. There are built in crash avoidance mechanisms in the drones themselves for safety so they do not crash, although they do get confused with a very brancy understory. They often miss smaller branches.If you look in the specifications of the particular UAV you will see they do not perform well with certain understories, so there is a chance of crashing. The same with telephone wires or other infrastructure that you have to be careful about. 

Also, it's good practice to always be able to see the drone, line-of-sight, which is actually a requirement for flight operations in many countries. Although you may be able to get around it by being in a tower or being in an open area. 

 Some studies have used AI classifiers and interesting frameworks to discuss full or partial detections, sometimes it is unknown if it is the animal of interest. I would carefully plan any fieldwork around the seasons and make sure to get any of your paperwork approved well before the months of the dry season. It's going to be your best chance to detect them. 

In my experience, ChatGPT-4 performs significantly better than version 3.5, especially in terms of contextual understanding. However, like any AI model, inaccuracies cannot be completely eliminated. I've also seen a video showing that Gemini appears to excel at literature reviews, though I haven't personally tested it yet. Here's the link to the video: https://www.youtube.com/watch?v=sPiOP_CB54A.

While GPT3.5 is good for some activities, GPT-4 and GPT4-turbo are much better. Anthropic Claude is also very good, on a par with GPT4 for many tasks. As someone else has mentioned, the key is in the prompt you use, though chatGPT is continually being extended to allow more contextual information to be included, for example external files that have been uploaded previously. Code execution and image generation are also possible with the paid version of chatGPT, and the latest models include data up to the end of 2023 (I think). You can also include calls to openAI or other APIs programatically to include these in your workflows for assisting with a variety of tasks.
Regarding end results - as always, we're responsible for whatever outputs are ultimately published/shared etc. 
For Conservation Evidence - you could try making your own GPT (chatGPT assistant) that can be published/shared using your own evidence base and prompt that should be well grounded and provide good responses (I should think). But don't use 3.5 for that, IMO.

Undoubted things will quickly evolve from just "straight" ChatGPTn, BARD, ClaudeAI, etc "standard" models, to more specialized Retrieval Augmentation Generation (RAG) , where facts from authoritative sources and rules are supplied as context for the LLM to summarize in its response. You can direct ChatGPT and BARD: "Your response must be based on the reference sections provided" up to a few K of tokens.  A huge amount of work is going into properly indexing reference materials in order to supply context to the reference models.  Folks like FAO and CGIAR are indexing all their agricultural knowledge to feed the standard ones with location, crop, livestock, etc specialty "knowledge" to provide farmers automated advice via mobile phones, etc. I can totally see the same for such mundane things as "how do I ... using ArcMAP or QGIS?" purely based on the vast amount of documentation and tutorials. Google, ChatGPT, etc do a really good job already; this is just totally focusing its response to the body of knowledge known in advance to be relevant.

I would highly recommend folks do some searching on  "LLM RAG" - that's what going nuts now across the board.

Then there's stuff  I like to call "un-SQL" ... unstructured query language .. that will take free-form queries to form SQL queries, with supporting visualization code.

see: 
"https://mlnotes.substack.com/p/no-more-text2sql-its-now-rag2sql"
"http://censusgpt.com" 

etc.

As far as writing and evaluating proposals, I saw a paper on how summarization of public review forms are being developed in several cities. 
see: "http://streetleveladvisors.com/?p=181562"

And that's just the standard LLMs; super-specialized LLMs based on Facebook Llama are being built purely based on domain-specific bodies of dialog - medical, etc.  LOTS of Phds to be done.

I think what will be critical in all this are strong audit trails and certification mechanisms to gain trust. Especially when it comes to deceptive simple terms like "best" 

Chris

Yes. The key output for synchronisation is the pulse per second (PPS) output which is synchronised very accurately to UTC. The TX from the GPS module is then useful for reading the time and positions. You generally don't need to be able to send commands to the module as most of the time the default settings are fine.

Also some other bat experts I'd recommend reaching out to, if you haven't already (for this and any bat acoustic questions) - Adrià López-Baucells, Nils Bouillard, Kate Jones, Stuart Newson, plus obviously anyone at the big orgs like Bat Conservation Int'l, etc. 

Hey @ccosma if you are interested in multiple cheap sensors,  Phil Atkin and I are making initial batches of pippyG bat detector. They can record 4-7 days at the moment, but could be modified to fit needs.

Same issues here. A MEMS is a great idea to pot, but you really need a piezoelectric element for this to work and not a MEMS based on capacitance (btw they're all capacitance, except for one now discontinued...). It was originally made by Vesper, but the company was bought out last year and the MEMS is no longer made. 

This is because you're no longer really doing a typical microphone, this would be a contact type hydrophone. For waterproofing, you can actually get a waterproof MEMS. As long as your not submerging this for an extended period, it should do the job. Be sure to keep the cable short between the PCB and the mic as you'll get noise as I've experienced. 

For generally answering your question on the "best" epoxy to with sound transparency, in general the harder the material the lower the acoustic impedance. I use Epotec 301 resin with a hardness of 85. Your shape will also influence the resonance frequencies, meaning the flat frequency response will now be distorted and you'll probably have distorted audio. . 

You generally don't want to pot MEMS microphones since they're designed to pick up on air pressure changes and adding any material in front of the microphone just introduces another transition layer where pressure waves need to propagate through. Also, potting the MEMS microphone can be tricky since if you get any material in the port, you could damage the microphone or drastically reduce its performance. If you want to seal something with epoxy, take a look at contact microphones. Higher frequencies will be attenuated but depending on the application, it could work. 

There are companies, however, that design fabrics that are waterproof/resistant but have a relatively low acoustic impedance. SAATI has a variety of samples that you can request and GORE makes Acoustic Vents that could work. You can design a mechanical housing around your MEMS microphone with small perforations that are covered by one of these materials. I did this for one of my latest projects and it holds up just fine in heavy rain conditions. 

Hey Sol, 

No problem at all. Depending on your configuration, the Audiomoth software would have to work on a PCB with an ESP32 chip which is the unit on the audiomoth/hydromoth, so you would have to make a PCB centered around this chip. You could mimic the functionality of the audiomoth software on another chip, like on a raspberry pi with python's pyaudio library for example. The problem you would have is that the H1A requires phantom power, so it's not plug and play. I'm not too aware with the H1e, but maybe you can control the microphone through the recorder that is programmable through activations by the RPi (not that this is the most efficient MCU for this application, but it is user friendly). A simpler solution might be to just record continuously and play a sound or take notes of when your 10 min deployment starts. I think it should last you >6 hours with a set of lithium energizer batteries. You may want to think about putting a penetrator on the PVC housing for a push button or switch to start when you deploy. They make a few waterproof options. 

Just somethign else that occured to me, but if you're dropping these systems, you'll want to ensure that the system isn't wobbling in the seagrass as that will probably be all you will hear on the recordings, especially if you plan to deploy shallower. For my studies in Curacao, we aim to be 5lbs negative, but this all depends on your current and surface action. You might also want to think about the time of day you're recording biodiversity in general. I may suggest recording the site for a bit (a couple days or a week) prior to your study to see what you should account for (e.g. tide flow/current/anthropogenic disturbance) and determine diel patterning of vocalizations you are aiming to collect if subsampling at 10 minutes. 

Cheers, 

Matt

Hi Sol,

If the maximum depth is 30m, it would be worth experimenting with HydroMoth in this application especially if the deployment time is short. As Matt says, the air-filed case means it is not possible to accurately calibrate the signal strength due to the directionality of the response. For some applications, this doesn't matter. For others, it may.

Another option for longer/deeper deployments would be an Aquarian H2D hydrophone which will plug directly into AudioMoth Dev or AudioMoth 1.2 (with the 3.5mm jack added). You can then use any appropriately sized battery pack.

If you also connect a magnetic switch, as per the GPS board, you can stop and start recording from outside the housing with the standard firmware.

Alex